检索结果-内蒙古大学图书馆

Audio Enhancement for Computer Audition—An Iterative Training Paradigm Using Sample Importance

Journal of Computer science & Technology 2024年第4期39卷 895-911页

作者： Manuel Milling Shuo Liu Andreas Triantafyllopoulos Ilhan Aslan Björn W.Schuller Chair of Embedded Intelligence for Health Care and Wellbeing University of AugsburgAugsburg 86159Germany Chair of Health Informatics München rechts der IsarTechnical University of MunichMunich 81675Germany Munich Center for Machine Learning Munich 80333Germany Huawei Technologies MunichMunich 80992Germany Munich Data Science Institute Garching 85748Germany Group on Language Audio and MusicImperial College LondonLondon SW72AZU.K.

Neural network models for audio tasks,such as automatic speech recognition(ASR)and acoustic scene classification(ASC),are susceptible to noise contamination for real-life *** improve audio quality,an enhancement module,which can be developed independently,is explicitly used at the front-end of the target audio *** this paper,we present an end-to-end learning solution to jointly optimise the models for audio enhancement(AE)and the subsequent *** guide the optimisation of the AE module towards a target application,and especially to overcome difficult samples,we make use of the sample-wise performance measure as an indication of sample *** experiments,we consider four representative applications to evaluate our training paradigm,i.e.,ASR,speech command recognition(SCR),speech emotion recognition(SER),and *** applications are associated with speech and nonspeech tasks concerning semantic and non-semantic features,transient and global information,and the experimental results indicate that our proposed approach can considerably boost the noise robustness of the models,especially at low signal-to-noise ratios,for a wide range of computer audition tasks in everyday-life noisy environments.

关键词： audio enhancement computer audition joint optimisation multi-task learning voice suppression

来源：评论

学校读者我要写书评

暂无评论

Differentiable and Scalable Generative Adversarial Models for data Imputation (Extended Abstract) 40

Differentiable and Scalable Generative Adversarial Models fo...

引用

40th IEEE International Conference on data Engineering, ICDE 2024

作者： Wu, Yangyang Wang, Jun Miao, Xiaoye Wang, Wenjia Yin, Jianwei Software College Zhejiang University Ningbo China Academy of Interdisciplinary Studies The Hong Kong University of Science and Technology Hong Kong Hong Kong Center for Data Science Zhejiang University Hangzhou China The State Key Lab of Brain-Machine Intelligence Zhejiang University Hangzhou China Guangzhou China College of Computer Science Zhejiang University Hangzhou China

ISBN: (纸本)9798350317152

The dramatically increasing volume of incomplete data makes the imputation models computationally infeasible in many real-life applications. In this paper, we propose an effective scalable imputation system named SCIS to significantly speed up the training of the differentiable generative adversarial imputation models under accuracy-guarantees for large-scale incomplete data. SCIS consists of two modules, differentiable imputation modeling (DIM) and sample size estimation (SSE). DIM leverages a new masking Sinkhorn divergence function to make an arbitrary generative adversarial imputation model differentiable, while for such a differentiable imputation model, SSE can estimate an appropriate sample size to ensure the user-specified imputation accuracy of the final model. Moreover, SCIS can also accelerate the autoencoder based imputation models. Extensive experiments upon several real-life large-scale datasets demonstrate that, our proposed system can accelerate the generative adversarial model training by 6.23x. Using around 1.27% samples, SCIS yields competitive accuracy with the state-of-the-art imputation methods in much shorter computation time. © 2024 IEEE.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

On the Effectiveness of Heterogeneous Ensemble Methods for Re-identification

arXiv

引用

arXiv 2024年

作者： Klüttermann, Simon Rutinowski, Jérôme Nguyen, Anh Grimme, Britta Roidl, Moritz Müller, Emmanuel TU Dortmund University Dortmund Germany Lamarr Institute for Machine Learning and Artificial Intelligence Germany Research Center Trustworthy Data Science and Security Germany

In this contribution, we introduce a novel ensemble method for the re-identification of industrial entities, using images of chipwood pallets and galvanized metal plates as dataset examples. Our algorithms replace commonly used, complex siamese neural networks with an ensemble of simplified, rudimentary models, providing wider applicability, especially in hardware-restricted scenarios. Each ensemble sub-model uses different types of extracted features of the given data as its input, allowing for the creation of effective ensembles in a fraction of the training duration needed for more complex state-of-the-art models. We reach state-of-the-art performance at our task, with a Rank-1 accuracy of over 77% and a Rank-10 accuracy of over 99%, and introduce five distinct feature extraction approaches, and study their combination using different ensemble methods. © 2024, CC BY.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Smaller Batches, Bigger Gains? Investigating the Impact of Batch Sizes on Reinforcement Learning Based Real-World Production Scheduling

Smaller Batches, Bigger Gains? Investigating the Impact of B...

引用

International Conference on Emerging Technologies and Factory Automation (ETFA)

作者： Arthur Müller Felix Grumbach Matthia Sabatelli Department of Machine Intelligence Fraunhofer IOSB-INA Lemgo Germany Center for Applied Data Science Hochschule Bielefeld Gütersloh Germany Department of Artificial Intelligence University of Groningen Groningen The Netherlands

ISBN: (数字)9798350361230

ISBN: (纸本)9798350361247

Production scheduling is an essential task in manufacturing, with Reinforcement Learning (RL) emerging as a key solution. In a previous work, RL was utilized to solve an extended permutation flow shop scheduling problem (PFSSP) for a real-world production line with two stages, linked by a central buffer. The RL agent was trained to sequence equallysized product batches to minimize setup efforts and idle times. However, the substantial impact caused by varying the size of these product batches has not yet been explored. In this follow-up study, we investigate the effects of varying batch sizes, exploring both the quality of solutions and the training dynamics of the RL agent. The results demonstrate that it is possible to methodically identify reasonable boundaries for the batch size. These boundaries are determined on one side by the increasing sample complexity associated with smaller batch sizes, and on the other side by the decreasing flexibility of the agent when dealing with larger batch sizes. This provides the practitioner the ability to make an informed decision regarding the selection of an appropriate batch size. Moreover, we introduce and investigate two new curriculum learning strategies to enable the training with small batch sizes. The findings of this work offer the potential for application in several industrial use cases with comparable scheduling problems.

关键词： Training Home appliances Job shop scheduling Production Reinforcement learning Manufacturing Complexity theory Manufacturing automation

来源：评论

学校读者我要写书评

暂无评论

Bayesian Inference of Transition Matrices from Incomplete Graph data with a Topological Prior

arXiv

引用

arXiv 2022年

作者： Perri, Vincenzo Petrović, Luka V. Scholtes, Ingo Data Analytics Group Department of Informatics University of Zurich Switzerland Machine Learning for Complex Networks Center for Artificial Intelligence and Data Science Julius-Maximilians-Universität Würzburg Germany

Many network analysis and graph learning techniques are based on discrete- or continuous-time models of random walks. To apply these methods, it is necessary to infer transition matrices that formalize the underlying stochastic process in an observed graph. For weighted graphs, where weighted edges capture observations of repeated interactions between nodes, it is common to estimate the entries of such transition matrices based on the (relative) weights of edges. However in real-world settings we are often confronted with incomplete data, which turns the construction of the transition matrix based on a weighted graph into an inference problem. Moreover, we often have access to additional information, which capture topological constraints of the system, i.e. which edges in a weighted graph are (theoretically) possible and which are not. Examples include transportation networks, where we may have access to a small sample of passenger trajectories as well as the physical topology of connections, or a limited set of observed social interactions with additional information on the underlying social structure. Combining these two different sources of information to reliably infer transition matrices from incomplete data on repeated interactions is an important open challenge, with severe implications for the reliability of downstream network analysis tasks. Addressing this issue, we show that including knowledge on such topological constraints can considerably improve the inference of transition matrices, especially in situations where we only have a small number of observed interactions. To this end, we derive an analytically tractable Bayesian method that uses repeated interactions and a topological prior to perform data-efficient inference of transition matrices. We compare our approach against commonly used frequentist and Bayesian approaches both in synthetic data and in five real-world datasets, and we find that our method recovers the transition probabilities with h

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Towards Development of Automated Knowledge Maps and databases for Materials Engineering using Large Language Models

arXiv

引用

arXiv 2024年

作者： Prasad, Deepak Pimpude, Mayur Alankar, Alankar Artificial Intelligence and Data Science Vivekanand Education Society Institute of Technology Mumbai India Department of Mechanical Engineering Indian Institute of Technology Bombay Mumbai400076 India Center for Machine Intelligence and Data Science Indian Institute of Technology Bombay Mumbai400076 India

In this work a Large Language Model (LLM) based workflow is presented that utilizes OpenAI ChatGPT model GPT-3.5-turbo-1106 and Google Gemini Pro model to create summary of text, data and images from research articles. It is demonstrated that by using a series of processing, the key information can be arranged in tabular form and knowledge graphs to capture underlying concepts. Our method offers efficiency and comprehension, enabling researchers to extract insights more effectively. Evaluation based on a diverse Scientific Paper Collection demonstrates our approach in facilitating discovery of knowledge. This work contributes to accelerated material design by smart literature review. The method has been tested based on various qualitative and quantitative measures of gathered information. The ChatGPT model achieved an F1 score of 0.40 for an exact match (ROUGE-1, ROUGE-2) but an impressive 0.479 for a relaxed match (ROUGE-L ,ROUGE-Lsum) structural data format in performance evaluation. The Google Gemini Pro outperforms ChatGPT with an F1 score of 0.50 for an exact match and 0.63 for a relaxed match. This method facilitates high-throughput development of a database relevant to materials informatics. For demonstration, an example of data extraction and knowledge graph formation based on a manuscript about a titanium alloy is discussed. © 2024, CC BY-NC-ND.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

CBAM-SAUNet: A novel attention U-Net for effective segmentation of corner cases 46

CBAM-SAUNet: A novel attention U-Net for effective segmentat...

引用

46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2024

作者： Rajamani, Srividya Tirunellai Rajamani, Kumar Angeline, J. Karthika, R. Schuller, Björn W. Embedded Intelligence for Health Care & Wellbeing University of Augsburg Germany Department of Artificial Intelligence Marwadi University Gujarat Rajkot India Department of Electronics and Communication Engineering Amrita School of Engineering Amrita Vishwa Vidyapeetham Coimbatore India Germany Munich Data Science Institute Germany Munich Center for Machine Learning GLAM-the Group on Language Audio & Music Imperial College London London United Kingdom

ISBN: (纸本)9798350371499

U-Net has been demonstrated to be effective for the task of medical image segmentation. Additionally, integrating attention mechanism into U-Net has been shown to yield significant benefits. The Shape Attentive U-Net (SAUNet) is one such recently proposed attention U-Net that also focuses on interpretability. Furthermore, recent research has focused on identification and reporting of corner cases in segmentation to accelerate the utilisation of deep learning models in clinical practise. However, achieving good model performance on such corner cases is a less-explored research area. In this paper, we propose CBAM-SAUNet which enhances the dual attention decoder block of SAUNet to improve its performance on corner cases. We achieve this by utilising a novel variant of the Convolutional Block Attention Module (CBAM)'s channel attention in the decoder block of SAUNet. We demonstrate the effectiveness of CBAM-SAUNet in the Automated Cardiac Diagnosis Challenge (ACDC) cardiac MRI segmentation challenge. Our proposed novel approach results in improvement in the Dice scores of 12% for Left Ventricle (LV) as well as Right Ventricle (RV) segmentation and 8% for Myocardium (MYO) for the identified corner-case dataset. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Influence of serotonin on the long-term muscle contraction of the Kohnstamm phenomenon

引用

Scientific Reports 2025年第1期15卷 1-11页

作者： Schmidt, Annika Meindl, Tobias Albu-Schäffer, Alin Franklin, David W. Stratmann, Philipp TUM School of Computation Information and Technology Technical University of Munich (TUM) Garching 85748 Germany Institute of Robotics and Mechtronics German Aerospace Center (DLR) Wessling 82234 Germany Munich Institute of Robotics and Machine Intelligence (MIRMI) Technical University of Munich (TUM) Munich 80992 Germany Department of Neurology University Hospital rechts der Isar Technical University of Munich Munich 81675 Germany Neuromuscular Diagnostics TUM School of Medicine and Health Technical University of Munich Munich 80992 Germany Munich Data Science Institute (MDSI) Technical University of Munich Munich 80992 Germany

Neuromodulation plays a central role in human movement control. An imbalance of neurotransmitters, especially dopamine and serotonin, can be associated with various neurological disorders causing tremors or spasms. Specifically, serotonin was shown to scale motoneuron excitability following intense muscle contractions, affecting short-latency reflexes. Likely, it may also influence motoneuron modulation in prolonged contractions, although this lacks experimental evidence. An intriguing test case for this hypothesis is presented by the Kohnstamm phenomenon, where sustained muscle contractions lead to prolonged amplified EMG activity and involuntary motions, aligning with the timescale of serotonergic amplification. The suspected serotonin influence on this effect was tested in a placebo-controlled human user study with 14 participants, where half were administered the serotonin antagonist Cyproheptadine and the other half a placebo. Comparing EMG and force responses after inducing the Kohnstamm phenomenon in the deltoid muscles revealed statistically significant faster EMG decay with the serotonin antagonist, while decay remained consistent in the placebo group compared to the response of the same participant group without medication. The force measurements showed the same trend, although no significance. This provides new data-based evidence that serotonin contributes to long-term motoneuron modulation, extending previous findings about the dedicated role and influence of this neurotransmitter. Additionally, the work suggests the phenomenon as an interesting test case to investigate the dedicated involvement of different neurocontrol mechanisms such as Persistent Inward Currents. © The Author(s) 2025.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Stable, fast and accurate: kernelized attention with relative positional encoding 21

Stable, fast and accurate: kernelized attention with relativ...

引用

Proceedings of the 35th International Conference on Neural Information Processing Systems

作者： Shengjie Luo Shanda Li Tianle Cai Di He Dinglan Peng Shuxin Zheng Guolin Ke Liwei Wang Tie-Yan Liu Center for Data Science Peking University Key Laboratory of Machine Perception MOE School of EECS Peking University Princeton University Microsoft Research University of Science and Technology of China Center for Data Science Peking University and Key Laboratory of Machine Perception MOE School of EECS Peking University and Institute for Artificial Intelligence Peking University

ISBN: (纸本)9781713845393

The attention module, which is a crucial component in Transformer, cannot scale efficiently to long sequences due to its quadratic complexity. Many works focus on approximating the dot-then-exponentiate softmax function in the original attention, leading to sub-quadratic or even linear-complexity Transformer architectures. However, we show that these methods cannot be applied to more powerful attention modules that go beyond the dot-then-exponentiate style, e.g., Transformers with relative positional encoding (RPE). Since in many state-of-the-art models, relative positional encoding is used as default, designing efficient Transformers that can incorporate RPE is appealing. In this paper, we propose a novel way to accelerate attention calculation for Transformers with RPE on top of the kernelized attention. Based upon the observation that relative positional encoding forms a Toeplitz matrix, we mathematically show that kernelized attention with RPE can be calculated efficiently using Fast Fourier Transform (FFT). With FFT, our method achieves O(n log n) time complexity. Interestingly, we further demonstrate that properly using relative positional encoding can mitigate the training instability problem of vanilla kernelized attention. On a wide range of tasks, we empirically show that our models can be trained from scratch without any optimization issues. The learned model performs better than many efficient Transformer variants and is faster than standard Transformer in the long-sequence regime.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning as an Improvement Heuristic for Real-World Production Scheduling

Reinforcement Learning as an Improvement Heuristic for Real-...

引用

International Conference on machine Learning and Applications (ICMLA)

作者： Arthur Müller Lukas Vollenkemper Department of Machine Intelligence Fraunhofer IOSB-INA Lemgo Germany Center for Applied Data Science Bielefeld University of Applied Sciences and Arts Gütersloh Germany

ISBN: (数字)9798350374889

ISBN: (纸本)9798350374896

The integration of Reinforcement Learning (RL) with heuristic methods is an emerging trend for solving optimization problems, which leverages RL's ability to learn from the data generated during the search process. One promising approach is to train an RL agent as an improvement heuristic, starting with a suboptimal solution that is iteratively improved by applying small changes. We apply this approach to a real-world multiobjective production scheduling problem. Our approach utilizes a network architecture that includes Transformer encoding to learn the relationships between jobs. Afterwards, a probability matrix is generated from which pairs of jobs are sampled and then swapped to improve the solution. We benchmarked our approach against other heuristics using real data from our industry partner, demonstrating its superior performance.

关键词： Industries Job shop scheduling Reinforcement learning Production Network architecture Transformers Search problems Market research Encoding Optimization

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：